Major shifts in three cancer risk factors over the past decade are changing the profile of women's cancer risks. The California Teachers Study (CTS), a prospective cohort study that has actively followed 133,479 female California public school professionals for incidence of cancer and other outcomes since 1995, is ideally poised to evaluate how the societal transitions of decreasing physical activity, increasing obesity, and increasing aspirin use affect risk of breast and other cancers. The CTS is a unique repository of exposure data over the life course on these important, modifiable cancer risk factors. Our three aims address associations between these shifting exposures and risk for breast cancer overall; breast cancer subtypes defined by hormone receptor status, HER-2 status, histology and stage; and other cancers, with subtypes defined by tumor location or histology. For breast cancer, gene-environment interactions will be assessed by genotyping single nucleotide polymorphisms (SNPs) in genes in key mechanistic pathways (particularly in inflammation, immune function, and insulin resistance pathways), as well as highly significant, replicated SNPs from genome-wide association studies. In Aim 1, we seek to understand better how physical activity reduces breast cancer risk, investigating whether variation in genes involved in fitness or athletic ability interacts with physical activity patterns defined by age and recency, and whether any associations with physical activity are modified by other environmental or genetic risk factors. We also examine how lifelong and changing patterns of body fat and obesity (Aim 2) and aspirin use (Aim 3), over a woman's lifespan influence women's breast, endometrial, ovarian or colon cancer risk overall and by subtype, investigating effect modification, and, for breast cancer, gene-environment interactions. To accomplish these aims, we will continue our current follow-up for cancer and other outcomes, which link to databases that capture 100% of these endpoints among California residents. We will collect and update time-dependent exposure data to broaden our existing information on key exposures from teenage years to old age. We will expand blood collection among breast cancer cases and controls, employing new collection approaches to ensure high participation and forging new scientific collaborations to fully utilize this biospecimen resource. We will continue analyzing key etiologic predictors of cancer using statistical approaches that account for missing data and secular changes in exposures and incorporate high-dimensional genetic data into existing models.